In a keynote address opening the prestigious annual Hot Chips symposium, AMD (NYSE: AMD) CTO Mark Papermaster delivered a vision for the coming "Surround Computing Era", and unveiled new processor architecture details, enabling technologies and design methodologies that will help drive the next era in computing. Surround computing is an extension of pervasive and ambient computing trends and describes an environment where computing technologies are completely natural and seamless parts of daily life.
"Surround computing imagines a world without keyboards or mice, where natural user interfaces based on voice and facial recognition redefine the PC experience, and where the cloud and clients collaborate to synthesize exabytes of image and natural language data. The ultimate goal is devices that deliver intelligent, relevant, contextual insight and value that improves consumers' everyday life in real time through a variety of futuristic applications. AMD is leading the quest for devices that understand and anticipate users' needs, are driven by natural user interfaces, and that disappear seamlessly into the background," said Papermaster during his opening remarks.
Papermaster explained that the Surround Computing Era will rely on robust "plug-and-play" IP portfolios including central processing units (CPUs), graphics processing units (GPUs), fixed function logic, and interconnect fabric. He also unveiled key details of AMD's upcoming "Steamroller" CPU architecture while underscoring the benefits of the industry-standard Heterogeneous Systems Architecture (HSA) that enables software developers to easily assign scalar and parallel compute workloads to the most appropriate compute units, and therefore optimize power.
"The road that leads us to the Surround Computing Era will be no less challenging and every bit as exciting as the 20-year journey in graphics processing that brought gamers from 'Pong' to today's modern game titles that feature stunning visual realism," Papermaster explained. "It will take an industry movement to complete this journey, and HSA provides the clear path forward to enable this next generation in computing."
Supporting Resources
- See video highlights of Mark Papermaster's Hot Chips keynote here (link will go live shortly after the keynote concludes).
- For more highlights of Mark Papermaster's Hot Chips keynote, as well as highlights from other AMD Hot Chips presenters, please read Mark Papermaster's Hot Chips blog here.
- See video highlights of AMD's presenters at Hot Chips 2012 here (link will go live the evening of August 29th)
Major news stories covered in the presentation:
1. "Steamroller" Architecture Details: Slides 14 to 17 outline the first look at "Steamroller", which is the core for the "Kaveri" APU among others. This is our first disclosure of "Steamroller" architectural details. We expect to see up to a 15% improvement in performance per W over the "Piledriver" core (those are design-level improvements rather than process level ones).
2. High Density (Thin) Libraries: See slide 22. For the products we're shipping today in 32nm, the AMD design team used a combination of automated place and route and hand-placed semi-custom design (top plot) which reduces power and area somewhat. In future products, AMD will be employing "dense" or "thin" libraries employed by our GPU design teams - but for CPU implementation - to deliver more power efficient computation. Both area and power are reduced by 30% (bottom plot), yielding a much more portable and energy efficiency CPU core employing industry standard design methodologies well adapted to a foundry model. So, essentially, we're able to use smart design to compensate for a full process node.
3. "Surround Computing": This is the big picture vision that Mark will be presenting and gives the industry a glimpse of where we're headed as a company and the kinds of experiences we'll be focused on delivering to consumers. Basically, we spent the last 10-20 years developing processors that could simulate visual reality, and we're going to spend the next 10-20 years turning that 'visual computing era' on its head and develop processors and platforms that start with an image (or series of images, or GPS data, or other environmental data), interpret their contents and context, and use that to deliver better real time experiences to users. This will leverage both cloud and client based processing, tying together all of the technologies and architectures we talk about in the presentation. Here's a simple example of how surround computing might look in the not-too-distant future. Imagine it's noon and you're heading out of the office for lunch. As you leave, your mobile device intuitively knows where you're going by automatically calculating multiple data points streaming in from the built-in GPS, clock, maps and cloud database connection. Since it knows your preferences so well, it goes to your favorite restaurant's website and checks the specials. A friendly automated voice inside your device springs to life and offers to take your lunch order. You reply, "lunch salad to go," and the device places the order for you. By the time you arrive at the restaurant, it's ready for pickup. An electronic payment was completed while you were in transit, so you don't even have to reach for your wallet. And all of this happens without downloading apps or typing on your phone.
AMD have a huge Hot Chips presence this year, in fact the biggest ever. In addition to Mark's keynote, AMD are presenting:
- Jeff Rupley will be discussing AMDs "Jaguar": A next generation low power x86 core
- Michael Mantor will be discussing AMD Radeon(TM) HD 7970 Graphics Core Next (GCN) Architecture
- Sebastian Nussbaum will be discussing the Second-Generation AMD A-Series APU Codename "Trinity" Architecture
- Bryan Black will be participating in a tutorial on Die Stacking
"Surround computing imagines a world without keyboards or mice, where natural user interfaces based on voice and facial recognition redefine the PC experience, and where the cloud and clients collaborate to synthesize exabytes of image and natural language data. The ultimate goal is devices that deliver intelligent, relevant, contextual insight and value that improves consumers' everyday life in real time through a variety of futuristic applications. AMD is leading the quest for devices that understand and anticipate users' needs, are driven by natural user interfaces, and that disappear seamlessly into the background," said Papermaster during his opening remarks.
Papermaster explained that the Surround Computing Era will rely on robust "plug-and-play" IP portfolios including central processing units (CPUs), graphics processing units (GPUs), fixed function logic, and interconnect fabric. He also unveiled key details of AMD's upcoming "Steamroller" CPU architecture while underscoring the benefits of the industry-standard Heterogeneous Systems Architecture (HSA) that enables software developers to easily assign scalar and parallel compute workloads to the most appropriate compute units, and therefore optimize power.
"The road that leads us to the Surround Computing Era will be no less challenging and every bit as exciting as the 20-year journey in graphics processing that brought gamers from 'Pong' to today's modern game titles that feature stunning visual realism," Papermaster explained. "It will take an industry movement to complete this journey, and HSA provides the clear path forward to enable this next generation in computing."
Supporting Resources
- See video highlights of Mark Papermaster's Hot Chips keynote here (link will go live shortly after the keynote concludes).
- For more highlights of Mark Papermaster's Hot Chips keynote, as well as highlights from other AMD Hot Chips presenters, please read Mark Papermaster's Hot Chips blog here.
- See video highlights of AMD's presenters at Hot Chips 2012 here (link will go live the evening of August 29th)
Major news stories covered in the presentation:
1. "Steamroller" Architecture Details: Slides 14 to 17 outline the first look at "Steamroller", which is the core for the "Kaveri" APU among others. This is our first disclosure of "Steamroller" architectural details. We expect to see up to a 15% improvement in performance per W over the "Piledriver" core (those are design-level improvements rather than process level ones).
2. High Density (Thin) Libraries: See slide 22. For the products we're shipping today in 32nm, the AMD design team used a combination of automated place and route and hand-placed semi-custom design (top plot) which reduces power and area somewhat. In future products, AMD will be employing "dense" or "thin" libraries employed by our GPU design teams - but for CPU implementation - to deliver more power efficient computation. Both area and power are reduced by 30% (bottom plot), yielding a much more portable and energy efficiency CPU core employing industry standard design methodologies well adapted to a foundry model. So, essentially, we're able to use smart design to compensate for a full process node.
3. "Surround Computing": This is the big picture vision that Mark will be presenting and gives the industry a glimpse of where we're headed as a company and the kinds of experiences we'll be focused on delivering to consumers. Basically, we spent the last 10-20 years developing processors that could simulate visual reality, and we're going to spend the next 10-20 years turning that 'visual computing era' on its head and develop processors and platforms that start with an image (or series of images, or GPS data, or other environmental data), interpret their contents and context, and use that to deliver better real time experiences to users. This will leverage both cloud and client based processing, tying together all of the technologies and architectures we talk about in the presentation. Here's a simple example of how surround computing might look in the not-too-distant future. Imagine it's noon and you're heading out of the office for lunch. As you leave, your mobile device intuitively knows where you're going by automatically calculating multiple data points streaming in from the built-in GPS, clock, maps and cloud database connection. Since it knows your preferences so well, it goes to your favorite restaurant's website and checks the specials. A friendly automated voice inside your device springs to life and offers to take your lunch order. You reply, "lunch salad to go," and the device places the order for you. By the time you arrive at the restaurant, it's ready for pickup. An electronic payment was completed while you were in transit, so you don't even have to reach for your wallet. And all of this happens without downloading apps or typing on your phone.
AMD have a huge Hot Chips presence this year, in fact the biggest ever. In addition to Mark's keynote, AMD are presenting:
- Jeff Rupley will be discussing AMDs "Jaguar": A next generation low power x86 core
- Michael Mantor will be discussing AMD Radeon(TM) HD 7970 Graphics Core Next (GCN) Architecture
- Sebastian Nussbaum will be discussing the Second-Generation AMD A-Series APU Codename "Trinity" Architecture
- Bryan Black will be participating in a tutorial on Die Stacking
